Experience

Author:

Al-Hussaini Leena1

Affiliation:

1. The University of Edinburgh

Abstract

Hunspell is a morphological spell checker and automatic corrector for Macintosh 10.6 and later versions. Aspell is a general spell checker and automatic corrector for the GNU operating system. In this experience article, we present a benchmarking study of the performance of Hunspell and Aspell. Ginger is a general grammatical spell checker that is used as a baseline to compare the performance of Hunspell and Aspell. A benchmark dataset was carefully selected to be a mixture of different error types at different word length levels. Further, the benchmarking data are from very bad spellers and will challenge any spell checker. The extensive study described in this work will characterize the respective softwares and benchmarking data from multiple perspectives and will consider many error statistics. Overall, Hunspell can correct 415/469 words and Aspell can correct 414/469 words. The baseline Ginger can correct 279/469 words. We recommend this dataset as the preferred benchmark dataset for evaluating newly developed “isolated word” spell checkers.

Publisher

Association for Computing Machinery (ACM)

Subject

Information Systems and Management,Information Systems

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Radiologic text correction for better machine understanding;Engineering Reports;2024-06-27

2. Formation of Text from Indian Sign Language using Convolutional Neural Networks;2022 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES);2022-07-15

3. Massive Text Normalization via an Efficient Randomized Algorithm;Proceedings of the ACM Web Conference 2022;2022-04-25

4. HINDIA: a deep-learning-based model for spell-checking of Hindi language;Neural Computing and Applications;2020-09-01

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3